Chapter K, Limit State of Shear Yielding K1-3

[MollyK]

I have been working on the connection designs for a building that uses LLRS design. We also have some built up box members where we need to check the box sections for concentrated forces. I am using Chapter K as my guide for checking these columns (I did not design the box sections, just the moment connections). I was trying to figure out the applicability of the punching shear limit state. So I have a situation where Bp=13in, B=18in and t=1.5in. (The diagram in the commentary of the 13th ed. is helpful).

A) The manual states "this limit need not be checked when Bp>(B-2t), nor when Bp<0.85B."
--> 13" is not greater than 15", however 13" is less than 15.3" (0.85*18").

B) Furthermore, when I checked the commentary, pg.16-368 in AISC 13th ed., I found "(0.85) <= beta <= (1-2t/B) Basis: punching shear failure"
--> Beta = Bp/B = 0.833, so in my situation I am outside of the inequality.

2 Questions:

In regards to A), if only one of the 2 criterion is met, does punching shear still need to be checked.

In regards to B), since I am outside of this inequality, does that mean I need to change my moment connections, the flange plate width so that I do fall within these 2 numbers?

In general, I understand the idea of punching shear, and I am really confused as to why there would even be a statement stating that this need not be checked when Bp<0.85B. I would think that a box section would be more likely to fail in punching shear the smaller your transverse plate becomes.

Any help would be appreciated, I just had an hour long conversation with the EOR, and neither of us understand why that 85% of B is in there to begin with.

 

[WillisV]

A) Punching needs to be checked if EITHER your Bp is greater than B-2t OR your Bp is less than 0.85B. So in your case, you do not have to check punching shear.

B) No, you don't have to change, you just don't have to check punching shear. The reason it is not checked for Bp < 0.85B is because you no longer get a shearing type failure for smaller transverse plates, instead you get a bending/yielding type failure which is covered by equation K1-2.

Think of the face of the HSS with the applied load as a concrete beam supported at each end by the HSS sidewalls with a concentrated load on it. When the load is almost directly over the "supports" (in this case the side walls), you do not need check shear as the load goes directly to the supports (hence the upper limit of the inequality); as the load moves out to just off of the support, shear dominates the behavior of the beam, thus in this case you have to check punching shear; as the load continues to move towards the middle of your "beam", flexure begins to dominate the behavior thus it is no longer necessary to check shear, and a bending type equation is more appropriate, in this case K1-2.

Nothing says you must not check shear, you are certainly welcome to check K1-3 if you wish, the code just states that it is not necessary to check because punching is no longer the dominate limit state at that point per the above discussion.

 

[WillisV]

Err...part A) should say needs NOT be checked...sorry about that.

 

[SteelPE]

You may want to look at AISC DG 24. This DG has some pretty useful charts and examples that help explain what/when limit states need to be checked.